Proc. Natl. Acad. Sci. USA Vol. 73, No. 12, pp. 4667-4671, December 1976
Medical Sciences
Hypophosphatemia: Mouse model for human familial hypophosphatemic (vitamin D-resistant) rickets (X-linkage/phosphate transport/animal model)
EVA M. EICHER*, JANICE L. SOUTHARD*, CHARLES R. SCRIVERt, AND FRANCIS H. GLORIEUXt The Jackson Laboratory, Bar Harbor, Maine 04609; t Departments of Biology and Pediatrics, McGill University, and the deBelle Laboratory for Biochemical Genetics, McGill University-Montreal Children's Hospital Research Institute, 2300 Tupper Street, Montreal, P.Q., Canada H3H IP3; and t The Genetics Unit, Shriners Hospital, 1529 Cedar Avenue, and Departments of Pediatrics and Experimental Surgery, Faculty of Medicine, McGill University, Montreal, Canada *
Communicated by Elizabeth S. Russell, September 27, 1976
ABSTRACT A new dominant mutation in the laboratory mouse, hypophosphatemia (gene symbol Hyp), has been identified. The Hyp gene is located on the X-chromosome and maps at the distal end. Mutant mice are characterized by hypophosphatemia, bone changes resembling rickets, diminished bone ash, dwarfism, and high fractional excretion of phosphate anion (low net tubular reabsorption). Phosphate supplementation of the diet from weaning prevents the appearance of severe skeletal abnormalities. The hypo hosphatemic male mouse resembles human males with X inked hypophosphatemia and the Hyp gene is presumably homologous with the X-linked human gene. The mouse model should facilitate study of the defect in transport of plasma inorganic phosphate anion.
Familial vitamin D-resistant rickets or X-linked hypophosphatemia (XLH) is characterized by X-linked dominant inheritance. Affected individuals have essentially normal serum calcium levels, hypophosphatemia, impaired net renal tubular reabsorption of phosphate anion, shortened stature, and vitamin D nonresponsive rickets or osteomalacia (1, 2). Although several hypotheses have been put forth, precise etiology of the disease remains unknown. We report the discovery of an X-linked dominant mutation named hypophosphatemia (gene symbol Hyp) in the laboratory mouse. Our findings indicate that the disease seen in the hypophosphatemic mouse is similar to XLH. Because both diseases are inherited as X-linked dominants, it is highly probable that the human and mouse diseases are caused by mutations affecting the homologous gene. Accordingly, the mouse should be a valuable model for elucidation of the basic defect in human XLH. MATERIALS AND METHODS Origin and Genetics. In 1966, six male mice with shortened trunk and hind limbs were noted in a linkage experiment at the Jackson Laboratory. By appropriate crosses, the new mutation was shown to be dominant and X-linked. Because the affected mice had a low serum phosphorus concentration, the mutation was named hypophosphatemia, gene symbol Hyp. Soon after its discovery, the mutant Hyp allele was transferred to the C57BL/6J inbred strain by repeated matings of Hyp/+ females to C57BL/6J +/Y males. All growth and physiological studies were conducted on mice of the C57BL/6J-Hyp strain. Diet. Unless otherwise noted, all mice were fed the mouse diet Old Guilford 96W containing 22.5% protein, 7.5% fat, 0.6% vitamin supplement, 0.22% calcium, and 0.74% phosphorus (wt/wt). The drinking water was acidified. Food and water were available ad libitum. Linkage. In order to determine the position of Hyp on the X chromosome, the following experiment was conducted. FeAbbreviation: XLH, X-linked hypophosphatemia.
4667
males that were + + Hyp/+ + + were mated to a Ta Bn +/Y male (Ta = tabby; Bn = bent-tail). The (Ta Bn +/+ + Hyp)Fl females were mated to + + +/Y CBA/J males. All offspring were classified for Ta (coat texture and color), Bn (bent, shortened tail), and Hyp (shortened hind limbs). Body Weight. Offspring produced by matings of Hyp/+ females to +/Y males were weighed at birth, toe-clipped for future identification, and weighed weekly until 43 days of age. The young were isolated from their parents at 21 days of age and separated by sex. Plasma Inorganic Phosphate and Calcium. Plasma inorganic phosphate (Pi) was determined by the method of Mattenheimer (3), modified for 50- to 100-Mil samples, on Hyp/+ and +/+ females and Hyp/Y and +/Y males ranging in age from 20 to over 400 days. Plasma calcium was measured with the Monitor calcium kit (Fisher Scientific Co.) on 35- to 300day-old individuals. Unless otherwise noted, plasma was collected between 10:00 a.m. and 1:00 p.m. Renal Excretion of Pi. Mice (10 Hyp/Y, 10 +/Y, 200 days of age) were housed in Gelman metabolic cages while urine was collected for 6 hr, in the morning, under fasting conditions. At the end of the collection period, blood was drawn by orbital sinus puncture. Serum was separated from the cells immediately and urine was acidified for phosphate measurement. Renal Cortex Pi. The Pi concentration in supernatants of 5% cold trichloroacetic acid homogenates of renal cortex was determined by the method of Vestergaard-Bogind (4). Effect of Diet Composition on Serum and Urine Pi. The effect of dietary calcium and phosphate on levels of serum and urine Pi was examined in two sets of eight Hyp/Y males and two sets of six +/Y males (240-270 days of age) fed one of two diets (described in Table 4) for 7 days before the experiment. Urine was collected as described above. Phosphate-Supplemented Drinking Water. A phosphatesupplemented diet ameliorates the rickets (or osteomalacia) in XLH human subjects (5). This therapy, therefore, was applied to hypophosphatemic male mice. Eight Hyp/Y and three +/Y males, 3-4 weeks of age, were given acidified drinking H20 ad libitum containing phosphate salts (Na2HPO4, 6.75 g and KH2PO4, 2.0 g per liter). Four Hyp/Y and three +/Y males of the same age were kept on acidified drinking H20. The mice were observed weekly by one of us (E.M.E.) for obvious changes in their hind limbs. Representatives of each genotype and treatment were killed after 11 or 18 weeks of phosphate therapy. Skeletons were prepared according to the method of Green (6), as modified by Eicher and Beamer (7). Other Measurements. Serum immune reactive parathyroid hormone was measured in three Hyp/Y and three +/Y males with CH-12M antiserum at the Shriners Hospital by standard immunoradioassay. The chicken antiserum, kindly provided by Dr. Claude Arnaud of the Mayo Clinic, senses COOH- and
4668
Proc. Natl. Acad. Sci. USA 73 (1976)
Medical Sciences: Eicher et al.
V.. t.
"-
A
B
C FIG. 1. Skeletal preparations from Hyp/Y and +/Y males. The left fore limbs and hind limbs of those mice in panels B and C were removed. (A) One-month-old HypIY (left) and +/Y littermate. (B) Five-month-old Hyp/Y male (left) and +/Y littermate. In both cases, the Hyp/Y male is significantly smaller than his +/Y littermate. Comparison of the Hyp/Y males in panels A and B reveals the kyphosis of the spine and rachitic rosary that develops as the disease progresses. (C) Two Hyp/Y males and one +/Y littermate (right) 4 months old. The Hyp/Y male in the center has been on a phosphate-supplemented diet since he was 1 month old. In this male there is no kyphosis of the spine, reduced rachitic rosary, and evidence of growth of the long bone (note femur) compared to his Hyp/Y male sibling maintained on a normal diet.
NH2-terminal fragments and the whole molecule of chicken parathyroid hormone, and it crossreacts with rodent and human parathyroid hormones. Bone ash weight and the calcium: phosphorus ratio were measured by standard methods. RESULTS General Phenotype. In matings of Hyp/+ female by +/Y male on the C57BL/6J inbred background, Hyp/+ and Hyp/Y individuals can be distinguished from their normal siblings at 21 days of age by their shortened hind limbs and tail (Fig. 1). On a hybrid background (observed as a consequence of the linkage experiments, see below), 10-20 more days are necessary in order to note a clear difference. Irrespective of background, the reduced body size persists throughout life. Kyphosis of the
thoracic vertebrae, rachitic rosary, and prominent bowing of the femur develop with age in mutant mice (Fig. 1). These skeletal abnormalities are more uniformly severe in the male compared to the female. Mutant animals may eventually develop extreme impairment of mobility from what appears to be a "locking" of the femur to the pelvic girdle. This condition did not appear to reduce the life span of hypophosphatemic males and females, since both survived to well over 2 years of age. Bone ash is reduced in Hyp/Y males (39.8% wt/wt) compared to +/Y siblings (59.1% wt/wt). In addition, the calcium:phosphorus ratio is similar in both (about 2:1). The values are the mean of triplicate determinations in 30-mg samples of bone from three male mice of each genotype. Hyp/+ females are fertile and raise their young. Not all
Medical Sciences: Eicher et al.
Proc. Natl. Acad. Sci. USA 73 (1976)
4669
Table 1. X-Linkage of Hyp Number of offspring Chromosome from 9 parent Bn Ta + + + Hyp + Ta + Bn + Hyp Bn Ta Hyp + + + + Ta Hyp + Bn +
9 47 54 22 9 14 34 3 3 Total 186
d
Total
31 69 8 1 14 23
78* 123 30 10* 28* 57 4 4* 334
1 1
148
g:a,0
4
Hyp/y
°C
Cross: Bn Ta +/+ + Hyp x + + +/Y 6. * One hundred twenty offspring used to calculate gene order and distance.
0
14
12
Hyp/Y males successfully sire offspring. Those that do may sire only one or two litters. Linkage. The X-linked loci Ta and Bn were used to locate the position of Hyp on the X chromosome. The offspring produced from the cross Bn Ta +/+ + Hyp female x + + +/Y male are given in Table 1. Because the Bn mutation does not have complete penetrance, only the animals known to receive the Bn rather than the + allele from their mother were considered in calculated gene order and distance. The order of loci as percentage recombination + standard error of the mean is: Bn-11.8 ± 2.9-Ta-26.7 + 4.0-Hyp. This places Hyp at the distal end of the mouse X chromosome. Growth. The mean weight values obtained for the hypophosphatemia females and males and their normal siblings are given in Fig. 2. The body weight of Hyp/Y males as compared to +/Y male siblings is significantly reduced by 8 days of age (P < 0.05) and remains so through 43 days of age. The weight of Hyp/+ females as compared to +/+ female siblings is significantly reduced by 22 days of age (P < 0.01) and remains so up to 43 days of age. Thus, the effect of the Hyp mutation is evident before weaning in the Hyp/Y males, and by weaning in Hyp/+ females. A single dose of the + allele in the Hyp/+ females does enhance growth up to weaning. A normal sexual dimorphism of body weight between normal female and male C57BL/6J mice is observed by 36 days of age (7); this finding was not observed in the hypophosphatemic females and males. Plasma Calcium and Phosphorus. Data for plasma calcium and Pi are given in Tables 2 and 3. The Hyp/+ females have
6
*
+/+
4
0o
Hyp/+
2
8
15
29
22
DAYS
OF
36
43
AGE
FIG. 2. Growth of hypophosphatemic mice and their normal sibling controls, 1-43 days of age. Open circles and squares are Hyp/+ and Hyp/Y individuals, respectively. Closed circles and squares are +/+ and +/Y individuals, respectively. Standard errors are shown for each point.
slightly but significantly lower plasma calcium than do +/+ females by 35-49 days of age (P < 0.05), and lower values continue to over 200 days of age. The same is true for Hyp/Y compared to +/Y males. Plasma Pi is significantly reduced in Hyp/+ compared to +/+ females (P < 0.05) and Hyp/Y compared to +/Y males (P < 0.05) by 20-49 days of age. These differences persist up to and beyond 400 days of age. Our data show that plasma Pi in C57BL/6J mice is higher during rapid growth that in mature adults, as it is in the growing human subject. Furthermore, the plasma Pi in mice appears to decline slowly between 20 and 400 days in +/+ and +/Y individuals. Although Hyp/Y males and Hyp/+ females do appear to have a decline in plasma Pi between 20 and 100 days, a continual decline after 100 days was not apparent. Phosphaturia and Urine Composition. Urinary Pi excretion
Table 2. Plasma phosphate levels (mg/100 ml) of hypophosphatemic and normal mice
Age in days* Sex
Genotype
20-49
50-99
100-199
200-299
300-399
400-750
9
Hyp/+
+/+
4.81 ± 0.14 (33) 8.11 + 0.23
4.47 ± 0.13 (27) 7.00 + 0.21
(23)
(22)
HyplY
4.13 + 0.09
3.77 + 0.12
3.99 + 0.30 (12) 6.03 ± 0.28 (10) 3.29 ± 0.17
3.68 + 0.28 (10) 6.25 + 0.39 (6) 3.36 + 0.27
4.21 + 0.14 (13) 6.15 + 0.22 (11) 3.60 + 0.17
3.97 + 0.12 (14) 5.83 + 0.60 (3) 4.26 + 0.73
+/Y
7.86 + 0.17 (25)
(43)
(25)
6.62 ± 0.26 (17)
(17)
6.96 ± -0.39 (9)
(5)
5.60 (1)
(20)
6.43 + 0.16 (14)
(5)
5.73 + 0.18 (4)
* Values are presented as mean + SEM. The values of Hyp/+ compared to +/+, and Hyp/Y compared to +/Y are significantly different (P < 0.05) for each age group by t-test for groups of unequal size. Numbers in parentheses are numbers of individuals tested.
Proc. Natl. Acad. Sci. USA 73 (1976)
Medical Sciences: Eicher et al.
4670
Table 4. Urine excretion of phosphate in relation to serum phosphate concentration and diet calcium and phosphate Diet composition (%, wt/wt) 1
20-
:~100-
EE
Urine Pi (excretion index)*
Calcium
Phosphate
+/Y
HyplY
+/Y Hyp/Y
0.22 0.72
0.74 0.67
6.2 + 0.6 7.6 + 0.9
2.4 +0.1 4.6 + 0.9
111 71
367 111
Litter mates were 240-270 days of age when placed on the diets. A total of eight Hyp/Y and six +/Y males were used. * Excretion index = (mg Pi mg-I creatinine in urine)/(mg Pi ml-' in serum) was calculated using pooled urine and plasma obtained from the mice placed in metabolic cages.
-
Hyp+20
Serum Pi (mg/100 ml, mean + SEM)
/Y Hyp/Y
FIG. 3. Urinary Pi excretion in relation to serum Pi in 10 +/Y and 10 Hyp/Y mice. The mice (200 days of age) were fed a diet containing 0.72% (wt/wt) phosphorus and 0.67% (wt/wt) calcium. Pi excretion is increased relative to serum Pi in HyplY animals. * Indicates statistical significance of finding (P < 0.001).
is similar in +/Y and Hyp/Y mice, when expressed as a coefficient of creatinine excretion (9). However, when Pi excretion is related to plasma Pi, the urinary excretion is significantly elevated in Hyp/Y mice (Fig. 3). Fractional excretion of phosphate in bladder urine, determined by an infusion method at endogenous plasma Pi in the anesthetized mouse, is 0.20 + 0.09 (mean ± SEM) in normal mice and 0.35 + 0.08 in Hyp/Y (P < 0.01) (three pairs of one +/Y and one Hyp/Y males were used). These values are comparable to fractional P1 excretion in bladder urine of normal and XLH human subjects, respectively. Urinary Pi excretion in relation to serum P1 remained elevated in Hyp/Y mice as compared to +/Y mice under various conditions of dietary phosphate and calcium intake (Table 4). Plasma Pi is diminished and urinary Pi excretion is increased at the usual dietary Ca:Pi ratio of 0.3. Hypophosphatemia and hyperphosphaturia are still apparent when the dietary Ca:Pi ratio is raised to 1.06. Since there is no histological evidence for parathyroid hyperplasia on the low-calcium diet, a change in cellular calcium in the kidney of Hyp/Y males may be a factor influencing the degree of phosphaturia. Calcium modulates the phosphaturia in XLH in man (9). Solutes other than Pi (e.g., amino acids or glucose) were not
present in excess in urine of Hyp/Y males compared to normal male siblings. Tissue Phosphate. Renal cortex Pi is 46.6 + 1.0 nmol/mg of protein (mean ± SEM) in eight +/Y mice and 46.6 + 1.1 nmol/mg of protein in six Hyp/Y individuals (no significant difference). This finding suggests that, while the tubular transport defect in Hyp/Y males compromises net reclamation of phosphate from the tubule lumen, it does not perturb the total cellular uptake of Pi anion from the combined peritubular and urinary surfaces of renal cortical epithelium. In fact, cellular Pi is greater than normal relative to plasma Pi in Hyp/Y males. Moreover, we have shown that renal cortex slices, which expose only the basolateral membranes of epithelial cells to the medium (10), accumulate 32P1 into organic and inorganic pools equivalently when prepared from kidneys of Hyp/Y and +/Y males (9). This finding suggests that the presumed Pi transport defect is not expressed significantly in the basolateral membrane under conditions of incubation in vtitro. Therapy. After 11 weeks on phosphate-supplemented drinking H20, growth of hind limbs in Hyp/Y males had occurred. In addition, kyphosis of the thoracic spine did not appear to develop. Skeletal preparations (Fig. 1) verified these visual observations and revealed that long bones, tail bones, and skull bones of Hyp/Y males were more normal, the rachitic changes were more reduced, and the spinal kyphosis was absent. No differences were noted in the skeletons on +/Y males kept on phosphate-supplemented drinking water compared to those on normal diet. Parathyroid Hormone and Gland Morphology. Parathyroid
Table 3. Plasma calcium levels (mg/100 ml) of hypophosphatemic and normal mice
Age in days*
Sex
Genotype
35-49
9
Hyp/+
8.44 + 0.12
d
50-99
9.09
+
0.09
100-199
200-475
8.92 + 0.09
9.19 + 0.13
+/+
9.03 ± 0.17
HyplY
(4)
(8) 9.79 ± 0.17 (7)
8.43 ± 0.12
8.64 ± 0.10
9.26 ± 0.11
9.38 + 0.15
(19)
(9)
+/Y
9.44 + 0.13
9.44 + 0.12
(5)
(12)
(10)
(7)
(7)
(3)
(16)
(12)
(15)
9.42 + 0.19
(6)
(10)
8.70 ± 0.09
8.46 ± 0.09
9.83 + 0.10
*Values are presented as mean ± SEM. The values for Hyp/+ compared to +/+ females Etnd Hjyp/Y compared to +/Y males are significantly different (P < 0.05) for each age group by t-test for groups of unequal size. Numbers in parentheses are numbers of individuals tested.
Medical Sciences: Eicher et al. glands were dissected from Hyp/Y and +/Y mice and examined by light microscopy. There was no evidence of parathyroid hyperplasia in Hyp/Y mice. Serum parathyroid hormone was not elevated in hypophosphatemic males (
Medical Sciences
Hypophosphatemia: Mouse model for human familial hypophosphatemic (vitamin D-resistant) rickets (X-linkage/phosphate transport/animal model)
EVA M. EICHER*, JANICE L. SOUTHARD*, CHARLES R. SCRIVERt, AND FRANCIS H. GLORIEUXt The Jackson Laboratory, Bar Harbor, Maine 04609; t Departments of Biology and Pediatrics, McGill University, and the deBelle Laboratory for Biochemical Genetics, McGill University-Montreal Children's Hospital Research Institute, 2300 Tupper Street, Montreal, P.Q., Canada H3H IP3; and t The Genetics Unit, Shriners Hospital, 1529 Cedar Avenue, and Departments of Pediatrics and Experimental Surgery, Faculty of Medicine, McGill University, Montreal, Canada *
Communicated by Elizabeth S. Russell, September 27, 1976
ABSTRACT A new dominant mutation in the laboratory mouse, hypophosphatemia (gene symbol Hyp), has been identified. The Hyp gene is located on the X-chromosome and maps at the distal end. Mutant mice are characterized by hypophosphatemia, bone changes resembling rickets, diminished bone ash, dwarfism, and high fractional excretion of phosphate anion (low net tubular reabsorption). Phosphate supplementation of the diet from weaning prevents the appearance of severe skeletal abnormalities. The hypo hosphatemic male mouse resembles human males with X inked hypophosphatemia and the Hyp gene is presumably homologous with the X-linked human gene. The mouse model should facilitate study of the defect in transport of plasma inorganic phosphate anion.
Familial vitamin D-resistant rickets or X-linked hypophosphatemia (XLH) is characterized by X-linked dominant inheritance. Affected individuals have essentially normal serum calcium levels, hypophosphatemia, impaired net renal tubular reabsorption of phosphate anion, shortened stature, and vitamin D nonresponsive rickets or osteomalacia (1, 2). Although several hypotheses have been put forth, precise etiology of the disease remains unknown. We report the discovery of an X-linked dominant mutation named hypophosphatemia (gene symbol Hyp) in the laboratory mouse. Our findings indicate that the disease seen in the hypophosphatemic mouse is similar to XLH. Because both diseases are inherited as X-linked dominants, it is highly probable that the human and mouse diseases are caused by mutations affecting the homologous gene. Accordingly, the mouse should be a valuable model for elucidation of the basic defect in human XLH. MATERIALS AND METHODS Origin and Genetics. In 1966, six male mice with shortened trunk and hind limbs were noted in a linkage experiment at the Jackson Laboratory. By appropriate crosses, the new mutation was shown to be dominant and X-linked. Because the affected mice had a low serum phosphorus concentration, the mutation was named hypophosphatemia, gene symbol Hyp. Soon after its discovery, the mutant Hyp allele was transferred to the C57BL/6J inbred strain by repeated matings of Hyp/+ females to C57BL/6J +/Y males. All growth and physiological studies were conducted on mice of the C57BL/6J-Hyp strain. Diet. Unless otherwise noted, all mice were fed the mouse diet Old Guilford 96W containing 22.5% protein, 7.5% fat, 0.6% vitamin supplement, 0.22% calcium, and 0.74% phosphorus (wt/wt). The drinking water was acidified. Food and water were available ad libitum. Linkage. In order to determine the position of Hyp on the X chromosome, the following experiment was conducted. FeAbbreviation: XLH, X-linked hypophosphatemia.
4667
males that were + + Hyp/+ + + were mated to a Ta Bn +/Y male (Ta = tabby; Bn = bent-tail). The (Ta Bn +/+ + Hyp)Fl females were mated to + + +/Y CBA/J males. All offspring were classified for Ta (coat texture and color), Bn (bent, shortened tail), and Hyp (shortened hind limbs). Body Weight. Offspring produced by matings of Hyp/+ females to +/Y males were weighed at birth, toe-clipped for future identification, and weighed weekly until 43 days of age. The young were isolated from their parents at 21 days of age and separated by sex. Plasma Inorganic Phosphate and Calcium. Plasma inorganic phosphate (Pi) was determined by the method of Mattenheimer (3), modified for 50- to 100-Mil samples, on Hyp/+ and +/+ females and Hyp/Y and +/Y males ranging in age from 20 to over 400 days. Plasma calcium was measured with the Monitor calcium kit (Fisher Scientific Co.) on 35- to 300day-old individuals. Unless otherwise noted, plasma was collected between 10:00 a.m. and 1:00 p.m. Renal Excretion of Pi. Mice (10 Hyp/Y, 10 +/Y, 200 days of age) were housed in Gelman metabolic cages while urine was collected for 6 hr, in the morning, under fasting conditions. At the end of the collection period, blood was drawn by orbital sinus puncture. Serum was separated from the cells immediately and urine was acidified for phosphate measurement. Renal Cortex Pi. The Pi concentration in supernatants of 5% cold trichloroacetic acid homogenates of renal cortex was determined by the method of Vestergaard-Bogind (4). Effect of Diet Composition on Serum and Urine Pi. The effect of dietary calcium and phosphate on levels of serum and urine Pi was examined in two sets of eight Hyp/Y males and two sets of six +/Y males (240-270 days of age) fed one of two diets (described in Table 4) for 7 days before the experiment. Urine was collected as described above. Phosphate-Supplemented Drinking Water. A phosphatesupplemented diet ameliorates the rickets (or osteomalacia) in XLH human subjects (5). This therapy, therefore, was applied to hypophosphatemic male mice. Eight Hyp/Y and three +/Y males, 3-4 weeks of age, were given acidified drinking H20 ad libitum containing phosphate salts (Na2HPO4, 6.75 g and KH2PO4, 2.0 g per liter). Four Hyp/Y and three +/Y males of the same age were kept on acidified drinking H20. The mice were observed weekly by one of us (E.M.E.) for obvious changes in their hind limbs. Representatives of each genotype and treatment were killed after 11 or 18 weeks of phosphate therapy. Skeletons were prepared according to the method of Green (6), as modified by Eicher and Beamer (7). Other Measurements. Serum immune reactive parathyroid hormone was measured in three Hyp/Y and three +/Y males with CH-12M antiserum at the Shriners Hospital by standard immunoradioassay. The chicken antiserum, kindly provided by Dr. Claude Arnaud of the Mayo Clinic, senses COOH- and
4668
Proc. Natl. Acad. Sci. USA 73 (1976)
Medical Sciences: Eicher et al.
V.. t.
"-
A
B
C FIG. 1. Skeletal preparations from Hyp/Y and +/Y males. The left fore limbs and hind limbs of those mice in panels B and C were removed. (A) One-month-old HypIY (left) and +/Y littermate. (B) Five-month-old Hyp/Y male (left) and +/Y littermate. In both cases, the Hyp/Y male is significantly smaller than his +/Y littermate. Comparison of the Hyp/Y males in panels A and B reveals the kyphosis of the spine and rachitic rosary that develops as the disease progresses. (C) Two Hyp/Y males and one +/Y littermate (right) 4 months old. The Hyp/Y male in the center has been on a phosphate-supplemented diet since he was 1 month old. In this male there is no kyphosis of the spine, reduced rachitic rosary, and evidence of growth of the long bone (note femur) compared to his Hyp/Y male sibling maintained on a normal diet.
NH2-terminal fragments and the whole molecule of chicken parathyroid hormone, and it crossreacts with rodent and human parathyroid hormones. Bone ash weight and the calcium: phosphorus ratio were measured by standard methods. RESULTS General Phenotype. In matings of Hyp/+ female by +/Y male on the C57BL/6J inbred background, Hyp/+ and Hyp/Y individuals can be distinguished from their normal siblings at 21 days of age by their shortened hind limbs and tail (Fig. 1). On a hybrid background (observed as a consequence of the linkage experiments, see below), 10-20 more days are necessary in order to note a clear difference. Irrespective of background, the reduced body size persists throughout life. Kyphosis of the
thoracic vertebrae, rachitic rosary, and prominent bowing of the femur develop with age in mutant mice (Fig. 1). These skeletal abnormalities are more uniformly severe in the male compared to the female. Mutant animals may eventually develop extreme impairment of mobility from what appears to be a "locking" of the femur to the pelvic girdle. This condition did not appear to reduce the life span of hypophosphatemic males and females, since both survived to well over 2 years of age. Bone ash is reduced in Hyp/Y males (39.8% wt/wt) compared to +/Y siblings (59.1% wt/wt). In addition, the calcium:phosphorus ratio is similar in both (about 2:1). The values are the mean of triplicate determinations in 30-mg samples of bone from three male mice of each genotype. Hyp/+ females are fertile and raise their young. Not all
Medical Sciences: Eicher et al.
Proc. Natl. Acad. Sci. USA 73 (1976)
4669
Table 1. X-Linkage of Hyp Number of offspring Chromosome from 9 parent Bn Ta + + + Hyp + Ta + Bn + Hyp Bn Ta Hyp + + + + Ta Hyp + Bn +
9 47 54 22 9 14 34 3 3 Total 186
d
Total
31 69 8 1 14 23
78* 123 30 10* 28* 57 4 4* 334
1 1
148
g:a,0
4
Hyp/y
°C
Cross: Bn Ta +/+ + Hyp x + + +/Y 6. * One hundred twenty offspring used to calculate gene order and distance.
0
14
12
Hyp/Y males successfully sire offspring. Those that do may sire only one or two litters. Linkage. The X-linked loci Ta and Bn were used to locate the position of Hyp on the X chromosome. The offspring produced from the cross Bn Ta +/+ + Hyp female x + + +/Y male are given in Table 1. Because the Bn mutation does not have complete penetrance, only the animals known to receive the Bn rather than the + allele from their mother were considered in calculated gene order and distance. The order of loci as percentage recombination + standard error of the mean is: Bn-11.8 ± 2.9-Ta-26.7 + 4.0-Hyp. This places Hyp at the distal end of the mouse X chromosome. Growth. The mean weight values obtained for the hypophosphatemia females and males and their normal siblings are given in Fig. 2. The body weight of Hyp/Y males as compared to +/Y male siblings is significantly reduced by 8 days of age (P < 0.05) and remains so through 43 days of age. The weight of Hyp/+ females as compared to +/+ female siblings is significantly reduced by 22 days of age (P < 0.01) and remains so up to 43 days of age. Thus, the effect of the Hyp mutation is evident before weaning in the Hyp/Y males, and by weaning in Hyp/+ females. A single dose of the + allele in the Hyp/+ females does enhance growth up to weaning. A normal sexual dimorphism of body weight between normal female and male C57BL/6J mice is observed by 36 days of age (7); this finding was not observed in the hypophosphatemic females and males. Plasma Calcium and Phosphorus. Data for plasma calcium and Pi are given in Tables 2 and 3. The Hyp/+ females have
6
*
+/+
4
0o
Hyp/+
2
8
15
29
22
DAYS
OF
36
43
AGE
FIG. 2. Growth of hypophosphatemic mice and their normal sibling controls, 1-43 days of age. Open circles and squares are Hyp/+ and Hyp/Y individuals, respectively. Closed circles and squares are +/+ and +/Y individuals, respectively. Standard errors are shown for each point.
slightly but significantly lower plasma calcium than do +/+ females by 35-49 days of age (P < 0.05), and lower values continue to over 200 days of age. The same is true for Hyp/Y compared to +/Y males. Plasma Pi is significantly reduced in Hyp/+ compared to +/+ females (P < 0.05) and Hyp/Y compared to +/Y males (P < 0.05) by 20-49 days of age. These differences persist up to and beyond 400 days of age. Our data show that plasma Pi in C57BL/6J mice is higher during rapid growth that in mature adults, as it is in the growing human subject. Furthermore, the plasma Pi in mice appears to decline slowly between 20 and 400 days in +/+ and +/Y individuals. Although Hyp/Y males and Hyp/+ females do appear to have a decline in plasma Pi between 20 and 100 days, a continual decline after 100 days was not apparent. Phosphaturia and Urine Composition. Urinary Pi excretion
Table 2. Plasma phosphate levels (mg/100 ml) of hypophosphatemic and normal mice
Age in days* Sex
Genotype
20-49
50-99
100-199
200-299
300-399
400-750
9
Hyp/+
+/+
4.81 ± 0.14 (33) 8.11 + 0.23
4.47 ± 0.13 (27) 7.00 + 0.21
(23)
(22)
HyplY
4.13 + 0.09
3.77 + 0.12
3.99 + 0.30 (12) 6.03 ± 0.28 (10) 3.29 ± 0.17
3.68 + 0.28 (10) 6.25 + 0.39 (6) 3.36 + 0.27
4.21 + 0.14 (13) 6.15 + 0.22 (11) 3.60 + 0.17
3.97 + 0.12 (14) 5.83 + 0.60 (3) 4.26 + 0.73
+/Y
7.86 + 0.17 (25)
(43)
(25)
6.62 ± 0.26 (17)
(17)
6.96 ± -0.39 (9)
(5)
5.60 (1)
(20)
6.43 + 0.16 (14)
(5)
5.73 + 0.18 (4)
* Values are presented as mean + SEM. The values of Hyp/+ compared to +/+, and Hyp/Y compared to +/Y are significantly different (P < 0.05) for each age group by t-test for groups of unequal size. Numbers in parentheses are numbers of individuals tested.
Proc. Natl. Acad. Sci. USA 73 (1976)
Medical Sciences: Eicher et al.
4670
Table 4. Urine excretion of phosphate in relation to serum phosphate concentration and diet calcium and phosphate Diet composition (%, wt/wt) 1
20-
:~100-
EE
Urine Pi (excretion index)*
Calcium
Phosphate
+/Y
HyplY
+/Y Hyp/Y
0.22 0.72
0.74 0.67
6.2 + 0.6 7.6 + 0.9
2.4 +0.1 4.6 + 0.9
111 71
367 111
Litter mates were 240-270 days of age when placed on the diets. A total of eight Hyp/Y and six +/Y males were used. * Excretion index = (mg Pi mg-I creatinine in urine)/(mg Pi ml-' in serum) was calculated using pooled urine and plasma obtained from the mice placed in metabolic cages.
-
Hyp+20
Serum Pi (mg/100 ml, mean + SEM)
/Y Hyp/Y
FIG. 3. Urinary Pi excretion in relation to serum Pi in 10 +/Y and 10 Hyp/Y mice. The mice (200 days of age) were fed a diet containing 0.72% (wt/wt) phosphorus and 0.67% (wt/wt) calcium. Pi excretion is increased relative to serum Pi in HyplY animals. * Indicates statistical significance of finding (P < 0.001).
is similar in +/Y and Hyp/Y mice, when expressed as a coefficient of creatinine excretion (9). However, when Pi excretion is related to plasma Pi, the urinary excretion is significantly elevated in Hyp/Y mice (Fig. 3). Fractional excretion of phosphate in bladder urine, determined by an infusion method at endogenous plasma Pi in the anesthetized mouse, is 0.20 + 0.09 (mean ± SEM) in normal mice and 0.35 + 0.08 in Hyp/Y (P < 0.01) (three pairs of one +/Y and one Hyp/Y males were used). These values are comparable to fractional P1 excretion in bladder urine of normal and XLH human subjects, respectively. Urinary Pi excretion in relation to serum P1 remained elevated in Hyp/Y mice as compared to +/Y mice under various conditions of dietary phosphate and calcium intake (Table 4). Plasma Pi is diminished and urinary Pi excretion is increased at the usual dietary Ca:Pi ratio of 0.3. Hypophosphatemia and hyperphosphaturia are still apparent when the dietary Ca:Pi ratio is raised to 1.06. Since there is no histological evidence for parathyroid hyperplasia on the low-calcium diet, a change in cellular calcium in the kidney of Hyp/Y males may be a factor influencing the degree of phosphaturia. Calcium modulates the phosphaturia in XLH in man (9). Solutes other than Pi (e.g., amino acids or glucose) were not
present in excess in urine of Hyp/Y males compared to normal male siblings. Tissue Phosphate. Renal cortex Pi is 46.6 + 1.0 nmol/mg of protein (mean ± SEM) in eight +/Y mice and 46.6 + 1.1 nmol/mg of protein in six Hyp/Y individuals (no significant difference). This finding suggests that, while the tubular transport defect in Hyp/Y males compromises net reclamation of phosphate from the tubule lumen, it does not perturb the total cellular uptake of Pi anion from the combined peritubular and urinary surfaces of renal cortical epithelium. In fact, cellular Pi is greater than normal relative to plasma Pi in Hyp/Y males. Moreover, we have shown that renal cortex slices, which expose only the basolateral membranes of epithelial cells to the medium (10), accumulate 32P1 into organic and inorganic pools equivalently when prepared from kidneys of Hyp/Y and +/Y males (9). This finding suggests that the presumed Pi transport defect is not expressed significantly in the basolateral membrane under conditions of incubation in vtitro. Therapy. After 11 weeks on phosphate-supplemented drinking H20, growth of hind limbs in Hyp/Y males had occurred. In addition, kyphosis of the thoracic spine did not appear to develop. Skeletal preparations (Fig. 1) verified these visual observations and revealed that long bones, tail bones, and skull bones of Hyp/Y males were more normal, the rachitic changes were more reduced, and the spinal kyphosis was absent. No differences were noted in the skeletons on +/Y males kept on phosphate-supplemented drinking water compared to those on normal diet. Parathyroid Hormone and Gland Morphology. Parathyroid
Table 3. Plasma calcium levels (mg/100 ml) of hypophosphatemic and normal mice
Age in days*
Sex
Genotype
35-49
9
Hyp/+
8.44 + 0.12
d
50-99
9.09
+
0.09
100-199
200-475
8.92 + 0.09
9.19 + 0.13
+/+
9.03 ± 0.17
HyplY
(4)
(8) 9.79 ± 0.17 (7)
8.43 ± 0.12
8.64 ± 0.10
9.26 ± 0.11
9.38 + 0.15
(19)
(9)
+/Y
9.44 + 0.13
9.44 + 0.12
(5)
(12)
(10)
(7)
(7)
(3)
(16)
(12)
(15)
9.42 + 0.19
(6)
(10)
8.70 ± 0.09
8.46 ± 0.09
9.83 + 0.10
*Values are presented as mean ± SEM. The values for Hyp/+ compared to +/+ females Etnd Hjyp/Y compared to +/Y males are significantly different (P < 0.05) for each age group by t-test for groups of unequal size. Numbers in parentheses are numbers of individuals tested.
Medical Sciences: Eicher et al. glands were dissected from Hyp/Y and +/Y mice and examined by light microscopy. There was no evidence of parathyroid hyperplasia in Hyp/Y mice. Serum parathyroid hormone was not elevated in hypophosphatemic males (
